This project is an extension of previous work directed toward the study of noninvasive methods to determine spatial and temporal relationships existing between tissues of clinical interest. The approach involves in vitro modeling of promising systems and the development of prototypes suitable for clinical evaluation. Recent work has focussed mostly on studies directed toward development of a versatile computerized dental radiographic system designed to be used both fluoroscopically and off-line to produce images which can be subtracted to show small changes in tissue occurring over long intervals of time, and combined in ways permitting tomosynthetic display of specific slices of the teeth and jaws. Progress continues in the development of both hard-ware and soft-ware for this system. Particularly notewothy is an automated technique recently created to estimate the projection angle of a radiograph relative to a set of standardized projections. The algorithm produces the correct solution from data characterized by angular discrepancies as large as 20 degree. Other software exploits differences in the mechanism of energy dissipation produced by hard and soft tissues enabling separation of these components for purposes of improved image analysis. More general research concerns decisions underlying the entire diagnostic process in dentistry. Recent data place in question prophylactic extraction of impacted third molars, and routine premedication with antibiotics of dental patients scheduled for surgery having a history of rheumatic fever.